bibtype |
C -
Conference Paper (international conference)
|
ARLID |
0447270 |
utime |
20240103210553.1 |
mtime |
20150925235959.9 |
SCOPUS |
84963972159 |
WOS |
000377943800441 |
DOI |
10.1109/EUSIPCO.2015.7362773 |
title
(primary) (eng) |
Adaptive approximate filtering of state-space models |
specification |
page_count |
5 s. |
media_type |
C |
|
serial |
ARLID |
cav_un_epca*0447269 |
ISBN |
978-0-9928626-4-0 |
ISSN |
2076-1465 |
title
|
Proceedings of 23rd European Signal Processing Conference |
page_num |
2236-2240 |
publisher |
place |
Nice |
name |
EURASIP |
year |
2015 |
|
|
keyword |
Approximate Bayesian computation |
keyword |
ABC |
keyword |
filtration |
author
(primary) |
ARLID |
cav_un_auth*0242543 |
full_dept (cz) |
Adaptivní systémy |
full_dept (eng) |
Department of Adaptive Systems |
department (cz) |
AS |
department (eng) |
AS |
full_dept |
Department of Adaptive Systems |
name1 |
Dedecius |
name2 |
Kamil |
institution |
UTIA-B |
fullinstit |
Ústav teorie informace a automatizace AV ČR, v. v. i. |
|
source |
|
cas_special |
project |
ARLID |
cav_un_auth*0303543 |
project_id |
GP14-06678P |
agency |
GA ČR |
country |
CZ |
|
abstract
(eng) |
Approximate Bayesian computation (ABC) filtration of state-space models replaces popular particle filters in cases where the observation models (i.e. likelihoods) are either computationally too demanding or completely intractable, but it is still possible to simulate from them. These sequential Monte Carlo methods evaluate importance weights based on the distance between the true observation and the simulated pseudo-observations. The paper proposes a new adaptive method consisting of probability kernel-based evaluation of importance weights with online determination of kernel scale. It is shown that the resulting algorithm achieves performance close to particle filters in the case of well-specified models, and outperforms generic particle filters and state-of-art ABC filters under heavy-tailed noise and model misspecification. |
action |
ARLID |
cav_un_auth*0319341 |
name |
23rd European Signal Processing Conference (EUSIPCO) |
dates |
31.08.2015-04.09.2015 |
place |
Nice |
country |
FR |
|
RIV |
BB |
reportyear |
2016 |
num_of_auth |
1 |
presentation_type |
PO |
inst_support |
RVO:67985556 |
permalink |
http://hdl.handle.net/11104/0249574 |
mrcbC62 |
1 |
confidential |
S |
arlyear |
2015 |
mrcbU14 |
84963972159 SCOPUS |
mrcbU34 |
000377943800441 WOS |
mrcbU63 |
cav_un_epca*0447269 Proceedings of 23rd European Signal Processing Conference 978-0-9928626-4-0 2076-1465 2236 2240 Nice EURASIP 2015 |
|